Researchers at the University of California, Los Angeles (UCLA) have developed a new AI model called SLIViT (SLice Integration by Vision Transformer) that can efficiently analyze and diagnose complex medical scans, such as MRIs and 3D retinal scans, with accuracy matching that of medical specialists. The model has wide adaptability across various imaging modalities and can analyze scans in a fraction of the time it takes human experts. This breakthrough has the potential to improve diagnostic efficiency, reduce data acquisition costs, and accelerate medical research.
- UCLA researchers have developed a deep-learning framework called SLIViT that can analyze and diagnose complex medical scans with high accuracy.
- SLIViT has wide adaptability across various imaging modalities, including 3D retinal scans, ultrasound videos, 3D MRI scans, and 3D CT scans.
- The model can analyze scans in a fraction of the time it takes human experts, reducing the time by a factor of 5,000.
- SLIViT overcomes the training dataset size bottleneck by leveraging prior “medical knowledge” from the more accessible 2D domain.
- The model has clinical applicability potential and can match the accuracy of manual expertise of clinical specialists.
- SLIViT is flexible and robust enough to work with clinical datasets that are not always in perfect order.
- The researchers plan to investigate how SLIViT can be leveraged for predictive disease forecasting and explore ways to ensure that systematic biases in AI models do not contribute to health disparities.